Hyaluronic acid (HA), the predominant glycosaminoglycan on malignant tumor cells, is bound to the cell surface by its receptor CD44. Levels of HA correlate with enhanced metastatic behavior, as do certain of the isoforms of CD44. Human breast cancer cell tumors grown in SCID mice decrease in size following i.v. hyaluronidase treatment, a 50% decrease occurring in four days. Brief exposure of cultured breast cancer cells in vitro to hyaluronidase eliminates not only the cell-associated HA, but also all of the malignant variants of CD44, leaving only two "benign" forms, CD44S and CD44E. It is now proposed to develop an animal model of oral squamous cell carcinoma (OSCC) using human cell lines grown orthotopically in nude mice, and to determine whether hyaluronidase can also cause tumor regression in this model. Simultaneously, coordinated studies using immunohistochemistry, FACS analyses, and RT-PCR will be performed to document modulation of HA and CD44 isoforms with hyaluronidase. It is postulated that the enzyme is a previously unrecognized anticancer therapeutic, that HA stabilizes its own receptor, and that eradication of HA by hyaluronidase causes down regulation of CD44 metastatic variant isoforms, thereby causing tumor regression, and prevention of further cancer spread. Furthermore, it is suspect that saliva creates an HA-rich environment in the oral cavity, and underlies the particularly aggressive course of OSCC's. In conjunction with these studies, levels and localization of HA, hyaluronidase, CD44, and its isoforms will be documented in an array of cultured OSCC cells of increasing aggressiveness, as well as in human biopsy specimens obtained from the oral cancer tissue and histopathology core. OSCC, comprising the vast majority of oral cancers, metastasizes widely. The degree of aggressiveness, unlike other human cancers, does not correlate with the histological appearance. Nor can it be predicted which patients with dysplasia or leukoplakia will go on to develop OSCC. The contention of this proposal is that the metabolism of HA is fundamental to the process of malignant progression, and that among the parameters measured in this proposal will be those that can function as predictive and prognostic indicators. Such potential markers will be of enormous clinical value in identifying not only patients at increased risk for developing malignancy, but also those with existing lesions that have particularly aggressive potential.